Cell-based protein expression is a well established method to obtain large amounts of a target protein. It enables accumulation and purification of the protein in quantities sufficient for various in vitro applications. Nonetheless, it is a tedious task to generate all the genetically modified organisms if a variety of proteins needs to be expressed. Additionally, purification of a distinct protein can not be performed according to a generalized protocol, but has to be optimized for every protein separately.
Cell-free expression represents a possibility to overcome many challenges of conventional protein expression in general and offers many advantages concerning our project in particular. Generally speaking, it saves a lot of time and money to avoid the generation of genetically modified organisms for every protein. Suitable DNA sequences are constructed once and can be stored until needed. In terms of purification, cell-free expression avoids the need for cell lysis preserving the protein from this harsh procedure. In cell-based expression, too strong induction often results in aggregating and therefore non-functional protein. This probability is minimized by using cell-free expression because the expressed protein is dispersed in a far larger volume than the intracellular space.
For copying DNA templates into protein microarrays in a microfluidic set-up, cell-free expression is the method of choice, because such a system is capable of expressing many different sequences at once. Additionally, the microfluidic system provides the opportunity to supplement the expression constantly with depleted components, like dNTPs, amino acids or energy sources like creatine phosphate, thus enabling high yields of protein.

We developed our own way of expressing proteins in a prokaryotic cell-free expression system based on an Escherichia coli lysate. Most established cell-free expression systems are based on rabbit reticulocytes, wheat germ cells, insect cells or E. coli cells. Due to the prokaryotic origin of most of the proteins expressed during our project, we decided to prepare our lysate based on an E. coli cell culture.
We used the E. coli strain BL21 for our lysate, carrying a gene coding for the T7 RNA polymerase which facilitates an improved expression of proteins during the cell-free expression. The cells were induced with IPTG prior to further treatment to initialize T7 polymerase expression.
Since all the coding sequences used in our project were manufactured synthetically, the E. coli BL21 cells were additionally transformed with a plasmid (pRARE2) containing the coding sequence of tRNAs for the translation of rare codons (AGA, AGG; AUA, CUA, GGA, CCC and CGG).
According to an optimized protocol (link) adapted from the European Molecular Biology Lab (EMBL) Heidelberg the lysate was prepared from the E. coli BL21 cells, frozen in liquid nitrogen and stored at -80°C. This lysate contains the whole transcription and translation apparatus of the cell, which allows expression of proteins from DNA templates.
However, the E. coli lysate is just one of the main components of the resulting cell-free expression mix. To be able to express proteins the system needs additional buffers, amino acids, nucleotides and energy sources. We obtained an L-amino acid kit from Sigma-Aldrich and prepared an accurate mix ourselves. The energy generating system based on creatine phosphate and phosphokinases and buffers (see protocol), nucleotides and amino acids were prepared and added to the lysate directly before expression.
The by then complete cell-free expression mix was subsequently used to express proteins. Therefore, the reaction was started by adding DNA coding for the respective proteins.

We compared our self-produced expression mix to commercially available kits to verify the functionality of our system. We could show (link to results) that the mix prepared with our own lysate produces proteins yields comparable to those obtained from commercial kits.

To validate expression of correctly folded proteins we used various methods like fluorescence measuring, luciferase assays and western blots. One of our first approaches was the expression of GFP, which can be detected via its fluorescence. For expression of GFP we used a plasmid template that was added to the self-made expression mix. Expression was then visualized via a microplate reader measurement that monitored the increase in fluorescence during progression of the expression. To verify the obtained data from this measurement a western blot was performed additionally. We used GFP antibodies to show the expression of GFP in our mix (link zu results).
As a second test for our system we performed a luciferase assay. In contrast to the expression of GFP which can be followed by a gradually rising of GFP over time, the luciferase assay is based on the emittance of light when luciferin is conversed into oxyluciferin. This reaction is catalyzed by the enzyme luciferase and doesn’t allow a tracking of expression over time. Moreover it is not dependent on correct folding and offers detection within seconds. We used the coding sequence of the firefly luciferase (pBESTluc) to express the protein with our cell-free expression system. The substrate luciferin was added to the mix after expression was finished and the luminescence was measured at once using a microplate reader (Link zu den results). This analysis helped us to evaluate the concentration of expressed protein in the cell-free expression system.

explanation of luciferase

Since we aim to use cell-free expression in the DiaCHIP we decided to validate the process in single steps. Starting with expression in a tube we spotted already expressed proteins on iRIF slides to check, whether or not we expressed detectable antigens that interacted with our antibodies at hand. As a next step we spotted our expression system on the slide without starting the reaction beforehand. By letting it incubate on a specific surface we could check for further possibilities. Lastly, we performed a cell-free expression from immobilized DNA in the finished setup of glass and PDMS slide.

To establish, whether our cell-free expression system reaches protein concentrations high enough to detect in the iRIf, we spotted readily cell-free expressed GFP on iRIF slides. This not only gave us information about the amount of protein, but also on its folding. By seeing an interaction with a corresponding antibody in the iRIf measurement, we could draw conclusions on the epitope.

Firstly we could detect wether the interaction between tagged proteins and the surface was strong enough to bind the proteins as soon as they were formed. This would show when measuring the slide in the iRIf afterwards. Secondly we wanted to assess if the surface was specific enough to not bind components of the mix itself.

This step would show us if we would already get our setup to work as a whole. Moreover, we could estimate, whether the diffusion patterns in a flow chamber differed too much from that in a suitable reaction tube.

Our biggest goal was to express antigens with the required tags in a concentration and folding state that suffices for an iRIf measurement. Here we focused on a C. tetani epitope flagged by a double His6 and a Spy-Tag (Spy-tag überhaupt erwähnen? Gute Frage, aber wir erwähnen den Spy-Catcher als Biobrick, oder? Dann macht das ja iwie schon Sinn den nochmal aufzugreifen) For cloning of this construct we used our own cloning site (link) and a vector that we optimized for cell-free expression.
The antigen was expressed in a tube reaction and afterwards spotted on a Ni-NTA surface. Sadly we could not show binding of the respective antibodies/serum with the iRIf measurement device. This could be due to folding problems of the antigen during cell-free expression or maybe the concentration of expressed C. tetani antigen was just not high enough to be detected by iRif measurement.

To ameliorate the quantity of the cell-free expressed proteins, we varied MgOAc and DNA template concentrations. Here we tested for optimal amounts by using luciferase DNA for direct detection.